Characterization of aromatic aminotransferases from Ephedra sinica Stapf

Article abstract of DOI:10.1007/s00726-015-2156-1

Ephedra sinica Stapf (Ephedraceae) is a broom-like shrub cultivated in arid regions of China, Korea and Japan. This plant accumulates large amounts of the ephedrine alkaloids in its aerial tissues. These analogs of amphetamine mimic the actions of adrenaline and stimulate the sympathetic nervous system. While much is known about their pharmacological properties, the mechanisms by which they are synthesized remain largely unknown. A functional genomics platform was established to investigate their biosynthesis. Candidate enzymes were obtained from an expressed sequence tag collection based on similarity to characterized enzymes with similar functions. Two aromatic aminotransferases, EsAroAT1 and EsAroAT2, were characterized. The results of quantitative reverse transcription-polymerase chain reaction indicated that both genes are expressed in young stem tissue, where ephedrine alkaloids are synthesized, and in mature stem tissue. Nickel affinity-purified recombinant EsAroAT1 exhibited higher catalytic activity and was more homogeneous than EsAroAT2 as determined by size-exclusion chromatography. EsAroAT1 was highly active as a tyrosine aminotransferase with α-ketoglutarate followed by α-ketomethylthiobutyrate and very low activity with phenylpyruvate. In the reverse direction, catalytic efficiency was similar for the formation of all three aromatic amino acids using l-glutamate. Neither enzyme accepted putative intermediates in the ephedrine alkaloid biosynthetic pathway, S-phenylacetylcarbinol or 1-phenylpropane-1,2-dione, as substrates.

Full text of DOI:10.1007/s00726-015-2156-1

Amino Acids (2016) 48:1209–1220
DOI 10.1007/s00726-015-2156-1
ORIGINAL ARTICLE
Characterization of aromatic aminotransferases from Ephedra
sinica Stapf
Korey Kilpatrick1,2 · Agnieszka Pajak2 · Jillian M. Hagel3 · Mark W. Sumarah2 ·
Efraim Lewinsohn4 · Peter J. Facchini3 · Frédéric Marsolais1,2
Received: 9 June 2015 / Accepted: 13 December 2015 / Published online: 1 February 2016
© Her Majesty the Queen in Right of Canada as represented by The Minister of Agriculture and Agri-Food 2016
Abstract Ephedra sinica Stapf (Ephedraceae) is a
broom-like shrub cultivated in arid regions of China, Korea
and Japan. This plant accumulates large amounts of the
ephedrine alkaloids in its aerial tissues. These analogs of
amphetamine mimic the actions of adrenaline and stimu-
late the sympathetic nervous system. While much is known
about their pharmacological properties, the mechanisms
by which they are synthesized remain largely unknown. A
functional genomics platform was established to investigate
their biosynthesis. Candidate enzymes were obtained from
an expressed sequence tag collection based on similarity
to characterized enzymes with similar functions. Two aro-
matic aminotransferases, EsAroAT1 and EsAroAT2, were
characterized. The results of quantitative reverse transcrip-
tion-polymerase chain reaction indicated that both genes
are expressed in young stem tissue, where ephedrine alka-
loids are synthesized, and in mature stem tissue. Nickel
affinity-purified recombinant EsAroAT1 exhibited higher
catalytic activity and was more homogeneous than EsA-
roAT2 as determined by size-exclusion chromatography.
EsAroAT1 was highly active as a tyrosine aminotransferase
with α-ketoglutarate followed by α-ketomethylthiobutyrate
and very low activity with phenylpyruvate. In the reverse
direction, catalytic efficiency was similar for the formation
of all three aromatic amino acids using l-glutamate. Nei-
ther enzyme accepted putative intermediates in the ephed-
rine alkaloid biosynthetic pathway, S-phenylacetylcarbinol
or 1-phenylpropane-1,2-dione, as substrates.
Keywords Ephedra sinensis Stapf Ephedraceae ·
Aminotransferase · Aromatic amino acids · l-Tyrosine ·
Pyridoxal 5′-phosphate · Transcriptome analysis
Introduction
Handling Editor: F. Hollfelder.
Ephedra sinica Stapf (Ephedraceae) is a perennial, broom-
like shrub cultivated in arid regions of China, Korea and
Japan (Bensky et al. 1986). This traditional Chinese medic-
inal plant has been utilized by humans for over 5000 years.
Ma Huang, a tea containing dried E. sinica stems was uti-
lized for the treatment of asthma, fever, coughs and lack
of sweating, and to promote urination and reduce edema
(Bensky et al. 1986; Bruneton 1995). All aerial tissues of
this plant accumulate the ephedrine alkaloids, with the
stems accumulating the highest amounts, up to 3 % of dry
weight (Krizevski et al. 2010).
Electronic supplementary material The online version of this
article (doi:10.1007/s00726-015-2156-1) contains supplementary
material, which is available to authorized users.
* Frédéric Marsolais
Frederic.Marsolais@agr.gc.ca
1
Department of Biology, University of Western Ontario,
London, ON N6A 5B7, Canada
2
Agriculture and Agri-Food Canada, Genomics
and Biotechnology, London Research and Development
Centre, 1391 Sandford St., London, ON N5V 4T3, Canada
The ephedrine alkaloids are a class of phenylpro-
pylamino alkaloids which are analogs of amphetamine. As
such they act as adrenergic agonists, mimicking the actions
of adrenaline by direct agonist activity and indirect release
of norepinephrine (Bruneton 1995; Rothman et al. 2003).
3
Department of Biological Sciences, University of Calgary,
Calgary, AB, Canada
4
Department of Aromatic, Medicinal and Spice Crops, Newe
Ya’ar Research Center, Agricultural Research Organization,
30095 Ramat Yishay, Israel
1 3

1210
K. Kilpatrick et al.
Fig. 1 Reactions catalyzed by E. sinica aromatic aminotransferases
These pharmacoactive stimulants, including (S)-cathi-
none, (1S,2S)-norpseudoephedrine, (1R,2S)-norephedrine,
(1S,2S)-pseudoephedrine and (1R,2S)-ephedrine are collec-
tively known as the ephedrine alkaloids. (1R,2S)-Norephed-
rine and (1R,2S)-ephedrine are used to treat acute asthma
attacks, rhinitis and sinusitis because of their vasoconstric-
tive properties (Bruneton 1995). (1R,2S)-Ephedrine is also
used to counteract anesthesia and overdoses of depressant
drugs as well as treatment for bronchial asthma and vari-
ous allergies (Lewis and Elvin-Lewis 1977). (1S,2S)-Pseu-
doephedrine is a widely used nasal decongestant (Bruneton
1995). (1R,2S)-Ephedrine is used as a controversial and
risky weight-loss or exercise supplement and a number of
deaths have been attributed to its use (Haller and Benow-
itz 2000). While much is known about the pharmacological
properties of these stimulants, the biological mechanisms
by which they are synthesized remain largely unknown.
The proposed biosynthetic pathway for production of
the ephedrine alkaloids can be found in recently published
articles by Hagel et al. (2012) and Krizevski et al. (2012).
Early radiolabelling experiments using Catha edulis (a tax-
onomically unrelated plant species which also produces the
ephedrine alkaloids) determined that l-phenylalanine (see
Fig. 1) is the initial precursor of the ephedrine alkaloids
(Leete 1958). Those findings were later confirmed using
Ephedra distachya stems (Yamasaki et al. 1969). So far,
only the step catalyzed by phenylalanine ammonia lyase
has been characterized at the molecular level in E. sinica
(Okada et al. 2008). An aromatic aminotransferase could
catalyze the formation of phenylpyruvate from l-phenyla-
lanine as a precursor to benzaldehyde (Hagel et al. 2011),
a pathway occurring in bacteria (Nierop Groot and de
Bont 1999). Benzaldehyde might then be condensed with
pyruvate by a benzaldehyde carboxyligase to form the
ephedrine alkaloid precursors (R)-phenylacetylcarbinol,
(S)-phenylacetylcarbinol,
(S)-2-hydroxypropiophenone
(R)-2-hydroxypropiophenone,
and
1-phenylpropane-
1,2-dione (Krizevski et al. 2012). Beside the provision of
precursors for the ephedrine biosynthetic pathway, ami-
notransferases may also be involved in the formation of the
intermediate (S)-cathinone from 1-phenylpropane-1,2-di-
one (Hagel et al. 2012; Grue-Sørensen and Spenser 1994).
Aminotransferases (EC 2.6.1) are enzymes that catalyze
the reversible transfer of an amino group from an amino
donor, typically an amino acid, to the carbonyl group of an
amino acceptor, typically a keto acid. They are involved in
a number of essential biological processes including anabo-
lism and catabolism of amino acids, nitrogen assimilation,
vitamin biosynthesis, plant stress responses and second-
ary metabolism (Givan 1980; Liepman and Olsen 2004).
Aminotransferases are part of the pyridoxal-5′-phosphate
dependent class of enzymes. Pyridoxal-5′-phosphate is
covalently bound through an imine linkage to the ε-amino
group of the conserved lysine residue in the active site of
the enzyme (Mehta et al. 1993). Most aminotransferases
are able to accept multiple substrates, given the usually
reversible nature of the reaction, and employ a ping–pong
reaction mechanism (Hirotsu et al. 2005). Among the seven
fold types of pyridoxal-5′-phosphate-dependent enzymes,
aminotransferases belong to types I and IV (Schneider
et al. 2000). There are six subfamilies of aminotransferases
defined by a different Pfam domain (Finn et al. 2014; Rudat
et al. 2012), five of which belong to fold-type I. Repre-
sentatives of subfamilies I and II are aspartate and aromatic
aminotransferases. ω-Aminotransferases belong to subfam-
ily III, and phosphoserine aminotransferase to subfamily
V. d-Alanine and branched chain amino acid aminotrans-
ferases belong to subfamily IV and to fold-type IV.
To investigate ephedrine alkaloid biosynthesis, a snap-
shot of the E. sinica stem transcriptome was generated by
1 3

Characterization of aromatic aminotransferases from Ephedra sinica Stapf
1211
next-generation sequencing of RNA extracted from young
stem tissue. The data were assembled and annotated to cre-
ate a functional genomics platform which was triaged for
candidate biosynthetic genes. This information was used to
identify and characterize two aromatic aminotransferases.
These data were mined for candidate aminotransferase
genes in the ephedrine alkaloid biosynthetic pathway based
on sequence similarity to previously characterized aromatic
aminotransferases.
RT‑PCR and cloning of E. sinica aromatic
aminotransferases
Materials and methods
Plant material
E. sinica RNA was treated with DNaseI (RNase-free)
(Life Technologies, Burlington, ON) following manufac-
turers’ guidelines to remove any residual DNA. The RNA
was reverse transcribed into cDNA using qScript cDNA
SuperMix (Quanta BioSciences, Gaithersburg, MD) fol-
lowing manufacturers’ guidelines and stored at −20 °C
until use. PCR primers used to amplify the open reading
frames examined in this study were as follows: forward
(5′-CTACGGATCCATGGCAAGCAATGGGGAATGG-3′)
and reverse (5′-CTACGGATCCATGGCAAGCAATGGG-
GAATGG-3′) for EsAroAT1 and forward (5′-CTACG-
GATCCATGGTGAAGTGGAATTTTAAGCCAAGC-3′)
and reverse (5′-CTACAAGCTTCTATAAACTCCTATAAT-
GTCTGCGACAGAATGC-3′) for EsAroAT2. Restriction
enzyme sites for BamHI (forward) and KpnI (reverse) are
underlined. Pfx50 DNA polymerase (Life Technologies)
was used and the PCR reaction consisted of denaturing at
94 °C for 2 min, 35 cycles of 94 °C for 15 s, 60.5 °C for
30 s and 68 °C for 90 s, and a final extension at 68 °C for
5 min. PCR products were cloned into pCR4Blunt-TOPO
vector and sequenced with a 3130 XL Genetic Analyzer
(Life Technologies). The candidate cDNAs were subcloned
into the pQE30 expression vector (Qiagen, Toronto, ON)
by restriction enzyme-mediated cloning using BamHI
and KpnI (EsAroAT1) or BamHI and HindIII (EsAroAT2)
restriction endonucleases (New England Biolabs, Whitby,
ON) and T4 DNA ligase (Life Technologies) following
manufacturers’ guidelines. E. coli strain XL10-Gold (Agi-
lent Technologies, Mississauga, ON) was used for cloning
and expression of candidate cDNAs.
E. sinica Stapf seeds were purchased from Horizon Herbs
(Williams, OR) and Richters Herbs (Goodwood, ON).
Seeds were sown in the spring and fall months in 50 %
sand, 50 % cactus soil mixture and grown in a greenhouse
with at least 14 h of light per day. Supplemental 220 V,
high-pressure sodium light and 20:20:20 (nitrogen–phos-
phorus–potassium) fertilizer was used. ‘Young’ E. sinica
tissues used in this study were the freshly grown tips of
the stems and light green in color. ‘Mature’ tissues were
the lower portions of stems and dark green in colour. After
sample collection, fresh plant material was immediately
frozen in liquid nitrogen and stored at −80 °C until use.
RNA extraction from E. sinica stem tissue
Total RNA was extracted from E. sinica stems following a
modified protocol for extracting RNA from tissues contain-
ing high levels of procyanidins (Wang and Vodkin 1994).
Approximately 500 mg of ground stem tissue was used in
place of 50 mg of lyophilized seed coats. After washing
with 2 M LiCl, the RNA pellet was resuspended in 100 μl
of diethylpyrocarbonate-treated water. Then, 10 μl of 3 M
NaOAc (pH 5.2) and 300 μl of 95 % (v/v) ethanol was
added and the solution was placed at −20 °C overnight.
The pellet was collected by centrifugation (25,000×g for
20 min at 4 °C) and washed twice with 70 % (v/v) ethanol.
The RNA pellet was dried, resuspended in diethylpyrocar-
bonate water and stored at −80 °C until use.
RT‑quantitative PCR
Next‑generation sequencing
E. sinica RNA was extracted in triplicate from both
young and mature stem tissue and cDNA was generated
as described above to determine relative gene expres-
sion levels of EsAroAT1 and EsAroAT2, compared to the
elongation factor 1-alpha (Ef1α) reference gene. PCR
primers used in RT-quantitative PCR analysis were as fol-
lows: forward (5′-GCTCTCCTTGCTTTCACCCT-3′)
and reverse (5′-TGGGATTTTGTCGGGGTTGT-3′) for
Ef1α (159 bp amplicon); forward (5′-TGACAGTTGGAT-
GCTCTCAGG-3′) and reverse (5′-CCTTCCACTCATC-
CTCTGGG-3′) for EsAroAT1 (174 bp amplicon); and
forward (5′-GCAGGCAAGAAGAGCTGTTTC-3′) and
A sample of total RNA (10 μl at 2.3 μg/μl) extracted
from young E. sinica stems was sent for next-generation
sequencing on the Illumina HiSeq 2000 RNA-seq plat-
form (single channel, 100 base pair, paired-end reads). The
sequencing was performed as per the manufacturers’ speci-
fications at the McGill University and Génome Québec
Innovation Centre. Expressed sequence tag data from the
library were assembled using the Velvet algorithms for de
novo short read assembly (Zerbino and Birney 2008) and
deposited in the MAGPIE (automated genomics project
investigation environment) portal (Groves et al. 2015).
1 3

1212
K. Kilpatrick et al.
reverse (5′-ACAAGGCTTCATACTGTGGGT-3′) for EsA-
roAT2 (184 bp amplicon). cDNA was diluted eightfold and
2 μl was amplified with 5 μl of SsoFast EvaGreen Super-
mix in a final volume of 10 μl using a CFX96 Real-Time
Detection System (Bio-Rad, Mississauga, ON). Reactions
were carried out in hard-shell 96-well clear PCR plates.
Controls without template were performed in triplicate in
each plate. Reactions were performed in triplicate and con-
tained primers at a concentration of 0.5 μM for EsAroAT1
and EsAroAT2, and 0.48 μM for Ef1α. The PCR program
consisted of an initial step of 95 °C for 3 min followed
by 35 cycles of 95 °C for 10 s and 59.5 °C for 30 s. Data
were analyzed with the CFX Manager 2.0 software and
expressed as the cycle number required to reach a threshold
fluorescence value (Ct). Data were normalized to the mean
Ct of the reference gene Ef1α, for which variation between
samples was ≤0.7. Melt curve analysis was used to confirm
the specificity of each primer pair in comparison with no
template controls. Efficiency of PCR reactions was calcu-
lated from a standard curve of Ct versus the logarithm of
starting template quantity. Each assay was optimized so
that the efficiency ranged between 100 and 113 %, with a
coefficient of determination (R2) ≥ 0.99.
8.0). Buffer exchange was performed using PD-10 columns
(GE Healthcare Life Sciences, Baie d’Urfé, QC) following
the manufacturers’ protocol. Recombinant proteins were
concentrated using an Amicon Ultra-4 filter (EMD Milli-
pore Corporation, Billerica, MA) and stored at −80 °C in
100 mM HEPES buffer (pH 8.2) with 1 mM sodium EDTA
and 20 % (v/v) glycerol. Eluted proteins were further puri-
fied using size-exclusion chromatography on a HiLoad
Superdex 200 prep grade column with 20 mM Tris buffer,
pH 7.5, 150 mM sodium chloride on an ÄKTApurifier sys-
tem (GE Healthcare Life Sciences). Molecular mass of
the eluted protein was calculated by generating a standard
curve of kav versus the logarithm of the molecular weight
of several protein standards, where kav = elution vol-
ume − void volume/column volume − void volume. Pro-
tein concentration was determined using the Bio-Rad pro-
tein assay solution with bovine serum albumin as standard.
SDS-PAGE was performed and gels stained with Coomas-
sie Brilliant Blue R250 stain to confirm that the proper pro-
tein had been purified by comparison of the empirical and
predicted molecular mass.
Absorption spectroscopy of EsAroAT proteins
Expression and purification of E. sinica aromatic
aminotransferases
Absorption spectra of purified recombinant proteins were
determined using a DU 600 spectrophotometer (Beckman
Coulter, Mississauga, ON). The typical wavelength scan
was between 250 and 600 nm and used 0.5 mg/ml of puri-
fied protein in 25 mM HEPES buffer with 100 mM sodium
chloride (pH 7.5).
Fresh overnight culture containing recombinant plasmid
were grown in LB media and used to inoculate 500 ml
of NZY media containing 100 μg/ml of ampicillin. Cul-
tures were incubated at 37 °C until an OD600 of ~0.6 was
reached. Expression of His-tagged candidate aminotrans-
ferases was induced by the addition of 1 mM isopropyl β-d-
1-thiogalactopyranoside and cells were cultured overnight
(18 h) at room temperature (~21 °C). Cells were collected
by centrifugation (5500×g for 5 min at 4 °C) and frozen at
−20 °C to enhance lysis. After native lysis buffer (50 mM
sodium phosphate monobasic, 300 mM sodium chloride
and 10 mM imidazole at pH 8.0) was used to resuspend
cells, 1 mg/ml lysozyme was added and lysates were placed
on ice for 30 min. French Press and sonication were used to
further lyse cells. After centrifugation to pellet cell debris
(15,500×g for 45 min at 4 °C), Ni–NTA agarose beads
(Qiagen) were added and the solutions were incubated at
4 °C for 2 h on a slow rotor. The Ni–NTA protein solu-
tions were centrifuged (3500×g for 3 min at 4 °C) and the
supernatant discarded. Pellets were washed ten times with
native wash buffer (50 mM sodium phosphate monobasic,
300 mM sodium chloride and 40 mM imidazole at pH 8.0)
to remove weakly bound proteins. Recombinant proteins
were eluted from the Ni–NTA agarose by the addition of
native elution buffer (50 mM sodium phosphate monoba-
sic, 300 mM sodium chloride and 250 mM imidazole at pH
Enzyme assays
Enzyme kinetics for the aromatic aminotransferase reac-
tions were determined by monitoring the absorbance
of reaction products by spectrophotometry as previ-
ously described (Collier and Kohlhaw 1972). Prior to
kinetic analysis, the enzymatic reactions described herein
were standardized by ensuring linear product forma-
tion at 10, 20, 30, 40, 50 and up to 60 min. Production
of 4-hydroxyphenylpyruvate was monitored at 331 nm
(ε331 = 19,920 M−1 cm−1), phenylpyruvate at 320 nm
(ε320 = 17,500 M−1 cm−1) and indole-3-pyruvate at 328 nm
(ε328 = 10,000 M−1 cm−1). The assay contained 100 mM
Tris–HCl buffer (pH 9.0), 0.1 μg recombinant protein
(0.01 μg for l-tyrosine), 0.1 mM sodium EDTA and the
following substrate concentrations: 0.3–6 mM amino
group donor (l-tyrosine, l-phenylalanine or l-tryptophan)
and 0.1–1.5 mM amino group acceptor (α-ketoglutarate)
in a total volume of 250 μl. Reactions were incubated at
30 °C for 30 min, terminated by the addition of 70 μl of
2 N NaOH and placed at room temperature for an addi-
tional 30 min before scanning, except for phenylpyruvate
1 3

Characterization of aromatic aminotransferases from Ephedra sinica Stapf
1213
which was scanned immediately. When required, pyridoxal
5′-phosphate was added at a concentration of 0.1 mM.
Enzyme kinetics for the aromatic aminotransferase reac-
tions with l-glutamate were determined by coupling with
α-ketoglutarate dehydrogenase and monitoring the absorb-
ance of NADH at 340 nm (ε340 = 6300 M−1 cm−1), as
previously described (Prabhu and Hudson 2010). Prior to
kinetic analysis, the reverse reaction was monitored con-
tinuously—at 3 min intervals between 3 and 30 min—and
demonstrated linear NADH formation for up to 30 min. The
500 μl assay contained 100 mM Tris–HCl buffer (pH 9.0),
2 μg recombinant protein, 0.1 mM sodium EDTA, 600 mg
α-ketoglutarate dehydrogenase (0.28 U/mg), 0.3 mM coen-
zyme A, 0.3 mM NAD+ and the following substrate con-
centrations: 0.3–25 mM amino group donor (l-glutamate)
and 0.05–0.9 mM 4-hydroxyphenylpyruvate, 0.05–0.9 mM
phenylpyruvate or 0.01–0.3 mM indole-3-pyruvate. Reac-
tions were incubated at 30 °C for 30 min then immediately
transferred to Greiner Bio One UV-Star 96-well, flat bottom
plates (Monroe, NC) and scanned using a Bio-Tek Power-
Wave XS plate reader (Winooski, VT). Enzymatic assays
excluding recombinant enzyme, excluding a substrate and
using heat-inactivated enzyme were used as negative con-
trols. Beer’s Law was used to determine concentrations of
the reaction products. Apparent Vmax and Km values were
calculated using Lineweaver–Burk (double reciprocal)
plots for analysis of the Michaelis–Menten equation.
Catalytic activity with 1,2-phenylpropanedione or (S)-
phenylacetylcarbinol was monitored in a similar way
using l-glutamate as amino group donor by coupling with
α-ketoglutarate dehydrogenase. Concentration of acceptor
substrate ranged between 0.01 and 1.0 mM. (S)-Phenylace-
tylcarbinol was purchased from Toronto Research Chemi-
cals (North York, Ontario, Canada) as an 80 % enantiomeric
excess mixture. Catalytic activity with 1,2-phenylpropan-
edione or (S)-phenylacetylcarbinol was also screened with
a mixture of amino acids as potential amino group donors
(Amino Acid Standard H) (Thermo Fisher Scientific, Burl-
ington, Ontario, Canada). Transamination of 1,2-phenylpro-
panedione or (S)-phenylacetylcarbinol was monitored by
GC–MS as described by Krizevski et al. (2010, 2012).
assembled and annotated by Dr. Christoph Sensen’s labo-
ratory at the University of Calgary. The assembly yielded
59,448 contigs spanning 89 million nucleotide bases
(Groves et al. 2015). This annotated expressed sequence
tag database was triaged to select candidate biosynthetic
enzymes involved in production of the ephedrine alkaloids.
Aromatic aminotransferase candidates
When blast searched with the deduced amino acid sequences
of prephenate aminotransferase from petunia, aromatic
aminotransferases from melon and rose, and tyrosine ami-
notransferase from opium poppy (Maeda et al. 2011; Gonda
et al. 2010; Hirata et al. 2012; Lee and Facchini 2011; Dal
Cin et al. 2011; de la Torre et al. 2006, 2014; Graindorge
et al. 2010), there were nine candidate contigs encoding aro-
matic aminotransferases that putatively catalyze reactions
involving aromatic substrates identified in the expressed
sequence tag database (ranging between 5 × 10−17 and 0 in
e value). Two were annotated as prephenate aromatic ami-
notransferases, and the remaining seven were annotated as
aromatic or tyrosine aminotransferases. Arabidopsis also
has seven annotated tyrosine aminotransferases (Prabhu
and Hudson 2010). RT-PCR was used to generate cDNAs
for each candidate, which were cloned into expression vec-
tors and expressed in Escherichia coli. Two of the candi-
dates, EsAroAT1 and EsAroAT2 (encoded by contigs 13244
and 12701, respectively) were found to have activity with
l-phenylalanine, l-tyrosine and l-tryptophan (Fig. 1) and
were selected for further characterization. A comparison
of their deduced amino acid sequences with aromatic ami-
notransferases from various organisms is summarized in
Table 1. A similar level of sequence identity of ca. 46 % was
observed between EsAroAT1 and the other plant aromatic
aminotransferases. There is often less than 50 % identity
when comparing aromatic aminotransferases within a spe-
cies or between different plant species (Prabhu and Hudson
2010). An alignment of the deduced amino acid sequences
revealed the presence of several conserved amino acid resi-
dues, including the invariant lysine which forms a covalent
intermediate with the co-factor, pyridoxal 5′-phosphate
(Online Resource 1).
Results
Gene expression analysis of EsAroAT1
Next‑generation sequencing and annotated E. sinica
expressed sequence tag library
Relative gene expression levels of EsAroAT1 and EsA-
roAT2 were determined by RT-quantitative PCR and com-
parison of their transcript levels to the elongation factor
1-alpha (Ef1α) reference gene in young and mature E.
sinica stem tissue (Table 2). EsAroAT1 expression was sim-
ilar between those two developmental stages, whereas EsA-
roAT2 expression was lower in mature stem as compared
with young stem by ca. fivefold.
E. sinica cDNA was generated and sequenced using the
Illumina HiSeq 2000 platform at the McGill University and
Génome Québec Innovation Centre. This high-through-
put sequencing method generated more than 95 million,
100 bp, paired-end reads. The collection of data were
1 3

1214
K. Kilpatrick et al.
Table 1 Identity in deduced
amino acid sequence of
EsAroAT2 and previously
characterized aromatic
aminotransferase enzymes from
plants and E. coli to EsAroAT1
Accession
Annotation
Organism
Percent identity
to EsAroAT1a
KC438923
Aromatic ATb
Ab-ArAT4
Ephedra sinica
46
48
47
47
46
46
45
44
20
KC954706.1
AB669189.1
KF511589.1
AT5G36160
AT5G53970
ADC45389.1
ADC33123.1
AAA24703.1
Atropa belladonna
Rosa ‘Yves Piaget’
Petunia hybrida
Phenylalanine AT
Phenylpyruvate AT
Tyrosine AT
TAT7
Arabidopsis thaliana
Arabidopsis thaliana
Cucumis melo
Aromatic AT
Tyrosine AT
Tyrosine AT
Papaver somniferum
Escherichia coli
a
Genbank accession number KC438292
b
AT aminotransferase
protein of ca. 68 kDa commonly co-purified with recom-
binant proteins from E. coli (Bolanos-Garcia and Davies
2006; Howell et al. 2006). This conclusion is supported by
the observation of a protein band of ca. 70 kDa in the cor-
responding size-exclusion fraction.
Table 2 Normalized gene expression level of EsAroAT1 and EsA-
roAT2 as determined by RT-quantitative PCR relative to Ef1α. Aver-
age Ct value is in parenthesis
Young stem
Mature stem
t test p value
EsAroAT1
EsAroAT2
1.30 0.10 (22.4) 1.02 0.18 (22.1) n. s.
1.47 0.28 (23.8) 0.29 0.06 (25.4) 0.002
Consistent with the visual observations and absorption
spectra, EsAroAT1 did not require exogenously added pyri-
doxal 5′-phosphate for aromatic aminotransferase activity
but EsAroAT2 did. Both EsAroAT1 and EsAroAT2 exhib-
ited tyrosine and phenylalanine aminotransferase activity in
initial enzymatic assays. However, the specific activity of
EsAroAT1 was ca. threefold higher than that of EsAroAT2
when using tyrosine as amino donor and ca. sevenfold
higher when using phenylalanine as donor (specific activi-
t test p value n.s.
0.002
Physical characteristics of recombinant EsAroAT
enzymes
The predicted molecular mass of EsAroAT1 is 45 and
46 kDa, respectively. Figure 2a shows an SDS-PAGE of
nickel affinity-purified recombinant EsAroAT1 and EsA-
roAT2. These preparations were used to acquire absorp-
tion spectra (Fig. 2b). The spectrum of EsAroAT1 had a
peak at approximately 430 nm, a characteristic of bound
pyridoxal 5′-phosphate (Ishikawa et al. 1996). The spec-
trum of EsAroAT2 did not have a peak in the 400 nm range
indicating that pyridoxal 5′-phosphate was not present.
These results confirmed visual observations that purified
EsAroAT1 was yellowish-green in colour, suggesting that
pyridoxal 5′-phosphate was bound to the enzyme. Similar
observations were reported for Arabidopsis tyrosine ami-
notransferase At5g36160 (Prabhu and Hudson 2010). The
molecular masses of native EsAroAT1 and EsAroAT2, as
determined by size-exclusion chromatography, were 75
and 71 kDa, respectively (Fig. 2c–e). These data confirmed
that EsAroAT1 and EsAroAT2 occur as homodimers, as
expected for aminotransferases. The EsAroAT2 preparation
was not as homogeneous as that of EsAroAT1 as demon-
strated by the presence of three distinct peaks in the chro-
matogram. These correspond to the monomeric form (ca.
42 kDa), the homodimer, and most likely a heterodimer
(ca. 117 kDa) formed with the chaperone DnaK, a host
ties of EsAroAT1 were equal to 7 and 2.1 nkatal mg−1
,
respectively). Neither EsAroAT1, EsAroAT2, nor the
remaining five candidate aromatic aminotransferases were
found to transaminate the intermediates of ephedrine alka-
loid biosynthesis, 1-phenylpropane-1,2-dione or S-phenyla-
cetylcarbinol when tested with l-glutamate, or a mixture of
proteinogenic amino acids as amino group donors. Due to
the lack of homogeneity of EsAroAT2, kinetic properties
were only characterized for EsAroAT1.
Kinetic characterization of EsAroAT1
After optimizing the temperature (30 °C), duration (30 min)
and amount of recombinant EsAroAT1 enzyme used (0.01–
0.1 μg) the optimum pH was determined for the forward
tyrosine aminotransferase reaction (Fig. 3a). There appeared
to be two values of pH optima, of 9.0 and 10.0. A high value
of optimal pH is a characteristic property of plant tyrosine
aminotransferases (De-Eknamkul and Ellis 1987a). There-
fore, plant aromatic aminotransferases are usually assayed
in this range of pH values. Since the highest activity was
observed with Tris–HCl buffer at pH 9.0, this was chosen for
enzymatic assays to determine apparent kinetic parameters.
1 3

Characterization of aromatic aminotransferases from Ephedra sinica Stapf
1215
Fig. 3 EsAroAT1 enzymatic activity. a pH optima of purified recom-
binant EsAroAT1. Forward tyrosine aminotransferase activity assays
were performed for 30 min at 30 °C in 100 mM of each buffer (Tris–
HCl, TAPS, CHES and CAPS) at the indicated pH values in the
presence of 0.05 μg of purified enzyme, 3 mM l-tyrosine (5), 1 mM
α-ketoglutarate (2) and 0.1 mM EDTA. Steady-state enzyme kinetics
of the forward (b) and reverse (c) reactions of purified recombinant
EsAroAT1. Values represent the average specific and error bars rep-
resent standard deviation (n = 3)
Plots of catalytic activity versus substrate concentra-
tion for the forward and reverse aromatic aminotransferase
reactions with α-ketoglutarate as amino group acceptor or
l-glutamic acid as amino group donor, respectively, are
compared in Fig. 3b, c. EsAroAT1 had the highest cata-
lytic activity with l-tyrosine as substrate in the forward
direction. In the reverse direction, phenylpyruvate and
4-hydroxyphenylpyruvate produced the highest activity.
The catalytic activity in the forward reaction with tyros-
ine was higher than for the reverse reaction, by ca. 50-fold.
Since it was recently reported that a cytosolic tyrosine
Fig. 2 Physical and spectral properties of recombinant EsAroAT1 and
EsAroAT2. SDS-PAGE of affinity-purified EsAroAT1 (a, left panel) and
EsAroAT2 (a, right panel). These preparations were used to acquire UV–
visible absorption spectra (b). The characteristic peak for bound pyridoxal
5′-phosphate at 433 nm was observed for EsAroAT1 (solid line) but not
for EsAroAT2 (dashed line). The same preparations were subjected to
size-exclusion chromatography on a Superdex 200 column. Elution pro-
file of EsAroAT1 (c) had a major peak corresponding to a homodimer
(first lane, e). Three peaks were observed for EsAroAT2 (d) correspond-
ing likely to a heterodimer with chaperone DnaK of ca. 68 kDa, a com-
mon host protein co-purified with heterologous recombinant proteins
from E. coli (lane 2, e), a homodimer (lane 3) and a monomer (lane 4)
1 3

1216
K. Kilpatrick et al.
Table 3 EsAroAT1 kinetic parameters for forward aromatic aminotransferase reactions
Substrate
Cosubstrate
Km (mM)
Vmax (×10−8 katal mg−1
)
kcat (s−1
)
kcat/Km (mM−1 s−1
)
l-Tyrosine
l-Phenylalanine
l-Tryptophan
l-Tyrosine
l-Tyrosine
l-Tyrosine
l-Tyrosine
LSD
α-Ketoglutarate (1 mM)
α-Ketoglutarate (1 mM)
α-Ketoglutarate (1 mM)
Phenylpyruvate (1 mM)
α-Ketomethylthiobutyrate (1 mM)
Oxaloacetate (1 mM)
0.91 0.04
8.67 1.21
5.37 1.04
0.698 0.054
0.372 0.011
0.271 0.021
0.258 0.026
1.06
70.3 1.8
18.0 1.8
7.56 0.37
0.239 0.003
5.90 0.14
1.73 0.06
0.543 0.021
1.69
32.8 0.8
36.1 0.8
8.39 0.82
3.53 0.17
0.112 0.002
2.76 0.06
0.807 0.029
0.254 0.010
0.793
0.972 0.043
0.672 0.111
0.161 0.011
7.41 0.05
2.99 0.19
0.988 0.060
0.523
Pyruvate (1 mM)
α-Ketoglutarate
l-Tyrosine (3 mM)
0.80 0.04
70.0 3.6
32.6 1.7
40.8 0.1
LSD Fisher’s protected least significant difference at p < 0.05
Table 4 EsAroAT1 kinetic parameters for reverse aromatic aminotransferase reactions
Substrate
Cosubstrate
Km (mM)
Vmax (×10−8 katal mg−1
)
kcat (s−1
)
kcat/Km (mM−1 s−1
)
Hydroxyphenylpyruvate
Phenylpyruvate
Indole-3-pyruvate
Hydroxyphenylpyruvate
LSD
l-glutamate (10 mM)
l-glutamate (10 mM)
l-glutamate (10 mM)
l-phenylalanine (10 mM)
0.105 0.018
0.115 0.006
0.037 0.009
0.311 0.024
0.0303
1.34 0.08
1.76 0.11
0.357 0.063
0.092 0.004
0.140
0.627 0.039
0.823 0.50
0.166 0.029
0.043 0.002
0.0660
6.06 0.930
7.18 0.46
4.51 0.432
0.139 0.008
1.06
l-glutamate
Phenylpyruvate (1 mM)
9.44 0.31
1.73 0.13
0.810 0.061
0.0858 0.0070
aminotransferase is involved in l-phenylalanine biosynthe-
sis in petunia (Yoo et al. 2013), and that a phenylalanine
:hydroxyphenylpyruvate aminotransferase (Ab-ArAT4) is
required for the biosynthesis of tropane alkaloids in Atropa
belladonna (Bedewitz et al. 2014), the corresponding reac-
tions were tested. The capacity of different organic acids to
substitute for α-ketoglutarate in the tyrosine aminotrans-
ferase reaction was also evaluated. Apparent kinetic param-
eters were determined using Lineweaver–Burk (double
reciprocal) plots to analyze the Michaelis–Menten equa-
tion. Tables 3 and 4 present the kinetic parameters for the
forward and reverse aromatic aminotransferase reactions,
respectively. In the forward direction with α-ketoglutarate
as acceptor substrate, the catabolism of l-tyrosine was
clearly the preferred reaction as it produced the highest
catalytic efficiency (kcat/Km) and reaction rate (Vmax). In the
reverse direction with l-glutamate as amino group donor,
the catalytic efficiencies were relatively similar for the
three aromatic aminotransferase reactions. Interestingly,
the Km values for the three organic acceptors using l-glu-
tamate as co-substrate were lower than that for l-tyrosine
in the forward direction. The Km value for indole-3-pyru-
vate was particularly low, being in the mid-micromolar
range (Table 4; Fig. 3b). Catalytic activity with l-tyrosine
and phenylpyruvate was low, with a ca. 300-fold lower
Vmax value and a ca. 200-fold lower value of catalytic effi-
ciency (kcat/Km) as compared with the preferred substrates
l-tyrosine and α-ketoglutarate. Similarly, for the reverse
reaction with hydroxyphenylpyruvate and l-phenylalanine,
the catalytic activity was lower than with hydroxyphe-
nylpyruvate and l-glutamate, with a ca. 14-fold reduced
Vmax value and a ca. 40-fold reduced value of catalytic
efficiency. When tested with l-tyrosine as the variable sub-
strate and a similar concentration of organic acid acceptor,
the highest Vmax value was observed with α-ketoglutarate,
followed by α-ketomethylthiobutyrate, oxaloacetate, pyru-
vate and phenylpyruvate.
Discussion
Aromatic aminotransferases (EC 2.6.1.57) catalyze reac-
tions using l-phenylalanine, l-tyrosine and l-tryptophan
as amino donors and typically use α-ketoglutarate as the
preferred amino acceptor in the forward direction (Fig. 1).
In the reverse direction, l-glutamate is typically the pre-
ferred amino donor and keto acid derivatives of the afore-
mentioned aromatic amino acids are amino acceptors.
The present results indicate that EsAroAT1 is most simi-
lar to previously characterized tyrosine aminotransferases.
Two Arabidopsis genes have been shown to encode cata-
lytically active tyrosine aminotransferases having TAIR
accession numbers At5g53970 (TAT7) (Holländer-Czytko
et al. 2005; Riewe et al. 2012), and At5g36160 (Prabhu
1 3

Characterization of aromatic aminotransferases from Ephedra sinica Stapf
1217
and Hudson 2010). Analysis of a T-DNA insertion mutant
for TAT7 revealed a role in the utilization of tyrosine for
the biosynthesis of tocopherols (Riewe et al. 2012). Both
tocopherols and plastoquinone, a co-factor of phytoene
desaturase involved in carotenoid biosynthesis, are derived
from the prenylquinone pathway (DellaPenna and Pogson
2006). In this pathway, tyrosine aminotransferase provides
4-hydroxyphenylpyruvate to hydroxyphenylpyruvate dehy-
drogenase which forms the common metabolic precursor,
homogentisate. Based on the crucial role of carotenoids in
protection against photooxidative damage (Lichtenthaler
2007), the reaction catalyzed by tyrosine aminotransferase
would be assumed to be essential for plant survival. Indeed,
tyrosine aminotransferase was shown to be targeted by
the herbicides cynmethylin and 5-benzyloxymethyl-1, 2
isoxazolines substituted with methylthiophene (methiozo-
lin) or pyridine (Grossmann et al. 2012). Not only was the
transamination of l-tyrosine being blocked in vivo, recom-
binant TAT7 was also inhibited in vitro by these chemicals.
Transgenics expressing a yeast prephenate dehydrogenase,
to by-pass tyrosine aminotransferase, exhibited tolerance to
one of these herbicides. Another tyrosine aminotransferase
was recently characterized from the opium poppy, Papa-
ver somniferum, with a role in benzylisoquinoline alkaloid
biosynthesis (Lee and Facchini 2011). Virus induced gene
silencing resulted in reduced levels of benzylisoquinoline
alkaloids, supporting the hypothesis that, like for TAT7,
the enzyme may function as a tyrosine aminotransferase
providing metabolic precursors for alkaloid biosynthesis.
Similarly, tyrosine aminotransferase has been implicated
in the biosynthesis of precursors for the formation of the
secondary metabolite rosmarinic acid in Anchusa officinalis
(De-Eknamkul and Ellis 1987b). As in the present study,
Riewe et al. (2012) reported a higher Vmax value of Arabi-
dopsis TAT7 for the forward reaction with l-tyrosine and
α-ketoglutarate versus the reverse reaction with 4-hydroxy-
phenylpyruvate and l-glutamate, by ca. 25-fold. However,
the value of kcat reported here for the forward reaction with
l-tyrosine and α-ketoglutarate is higher than those previ-
ously reported for plant tyrosine aminotransferases, by ca.
10–100 fold (Riewe et al. 2012; Lee and Facchini 2011;
Prabhu and Hudson 2010; Yoo et al. 2013). The kinetic
parameters reported for the forward reactions are similar
to those determined for mouse tyrosine aminotransferase
(Mehere et al. 2010). Among plant aromatic aminotrans-
ferases, only the recently characterized phenylalanine:hyd
roxyphenylpyruvate aminotransferase from Atropa bella-
donna has similar kinetic parameters as EsAroAT1, with a
kcat/Km value of 87.2 mM−1 s−1 with 4-hydroxyphenylpyru-
vate and l-phenylalanine as substrates (Bedewitz et al.
2014).
prephenate is transaminated to arogenate, which is con-
verted to l-phenylalanine by arogenate dehydratase
(Maeda and Dudareva 2012; Tohge et al. 2013; Maeda
et al. 2010; Cho et al. 2007; Corea et al. 2012). In an alter-
native pathway, prephenate is converted to phenylpyru-
vate by prephenate dehydratase, and phenylpyruvate is
transaminated to l-phenylalanine. In transgenic Arabidop-
sis expressing a bacterial bifunctional chorismate mutase/
prephenate dehydratase, this pathway appeared to be
stimulated (Tzin et al. 2009). The occurrence of this alter-
native pathway in plants is supported by the functional
complementation of the yeast prephenate dehydratase
pha2 mutant by selected Arabidopsis arogenate dehy-
dratase genes (Bross et al. 2011). Recent evidence impli-
cates a phenylpyruvate aminotransferase using l-tyrosine
as an amino group donor to form l-phenylalanine (Yoo
et al. 2013). The present data indicate that, unlike this
tyrosine:phenylpyruvate aminotransferase and the pheny
lalanine:hydroxyphenylpyruvate aminotransferase from
Atropa belladonna (Ab-ArAT4), EsAroAT1 has very low
catalytic activity with a pair of aromatic substrates.
When assaying different organic acid accep-
tors at the same concentration, using l-tyrosine as
the variable substrate, the second highest Vmax value
after that with α-ketoglutarate was observed with
α-ketomethylthiobutyrate. This compound is an interme-
diate in the l-methionine salvage pathway. In mammals, it
is transaminated to l-methionine by glutamine transami-
nases K and L [E.C. 2.6.1.‒] (Cooper 2004; Pinto et al.
2014). Recent research has identified the correspond-
ing gene and enzyme in Bacillus subtilis and tomato and
maize, whose activity is coupled with an ω-amidase,
hydrolyzing α-ketoglutaramate produced in the reaction
(Zhang and Marsolais 2014; Ellens et al. 2014). Knock-
out mutants of glutamine transaminase K in B. subtilis are
viable, suggesting that other aminotransferases may sub-
stitute for its function. In Klebsiella pneumoniae, tyros-
ine aminotransferase is involved in the transamination of
α-ketomethylthiobutyrate (Heilbronn et al. 1999). The pre-
sent data indicate that plant tyrosine aminotransferases may
share the capacity to transaminate this substrate.
According to the present results, EsAroAT1 is most
likely to function as a tyrosine aminotransferase in the pre-
nylquinone pathway, an essential enzyme in higher plants
(Riewe et al. 2012; Grossmann et al. 2012). EsAroAT1 is
unlikely to contribute to ephedrine alkaloid biosynthesis
through the catabolism of l-phenylalanine to phenylpyru-
vate as a precursor to benzaldehyde, since the Km value
for l-phenylalanine was nearly tenfold higher than that for
l-tyrosine with α-ketoglutarate as co-substrate. Neither is
EsAroAT1 likely to provide l-phenylalanine as a precur-
sor to ephedrine alkaloids by acting as a phenylpyruvate
aminotransferase. Although l-phenylalanine formation
In plants, l-phenylalanine biosynthesis can occur
predominantly via the arogenate pathway, whereby
1 3

1218
K. Kilpatrick et al.
represents its preferred reverse reaction, the enzyme has a
clear preference for the forward reaction with l-tyrosine. In
addition, l-tyrosine is implicated as the amino group donor
in l-phenylalanine biosynthesis, at least in dicotyledonous
plants (Yoo et al. 2013).
References
Bedewitz MA, Gongora-Castillo E, Uebler JB, Gonzales-Vigil E,
Wiegert-Rininger KE, Childs KL, Hamilton JP, Vaillancourt B,
Yeo YS, Chappell J, DellaPenna D, Jones AD, Buell CR, Barry
CS (2014) A root-expressed l-phenylalanine:4-hydroxyphe-
nylpyruvate aminotransferase is required for tropane alkaloid
biosynthesis in Atropa belladonna. Plant Cell 26:3745–3762.
doi:10.1105/tpc.114.130534
The lack of requirement for exogenous pyridoxal
5′-phosphate has been reported for numerous plant aro-
matic aminotransferases, for example by Gonda et al.
(2010), Prabhu and Hudson (2010), Hirata et al. (2012),
and Lee and Facchini (2011) and references therein. The
reason for the lack of co-purification of EsAroAT2 with
its co-factor, pyridoxal 5′-phosphate, is not obvious. In
the crystal structure of human tyrosine aminotransferase
(PDB ID 3DYD), the co-factor binding site, including the
conserved arginine interacting with the phosphate group,
corresponding to Arg288 in mouse tyrosine aminotrans-
ferase (Mehere et al. 2010), is very close to the dimer
interface. It is therefore possible that the lack of co-factor
may result in a weaker dimer, hence the different molecu-
lar forms of EsAroAT2. Alternatively, the absence of the
co-factor may have global effects of the structure of the
protein.
Bensky D, Gamble A, Kaptchuk T (1986) Chinese Herbal Medicine.
Materia Medica, Eastland, Seattle, Washington
Bolanos-Garcia VM, Davies OR (2006) Structural analysis and classi-
fication of native proteins from E. coli commonly co-purified by
immobilised metal affinity chromatography. Biochim Biophys
Acta 1760:1304–1313. doi:10.1016/j.bbagen.2006.03.027
Bross CD, Corea ORA, Kaldis A, Menassa R, Bernards MA, Kohalmi
SE (2011) Complementation of the pha2 yeast mutant sug-
gests functional differences for arogenate dehydratases from
Arabidopsis thaliana. Plant Physiol Biochem 49:882–890.
doi:10.1016/j.plaphy.2011.02.010
Bruneton J (1995) Pharmacognosy, phytochemistry. Medicinal Plants.
Intercept Ltd., Hampshire
Cho MH, Corea ORA, Yang H, Bedgar DL, Laskar DD, Anterola
AM, Moog-Anterola FA, Hood RL, Kohalmi SE, Bernards MA,
Kang C, Davin LB, Lewis NG (2007) Phenylalanine biosyn-
thesis in Arabidopsis thaliana—identification and characteriza-
tion of arogenate dehydratases. J Biol Chem 282:30827–30835.
doi:10.1074/jbc.M702662200
In conclusion, a highly active tyrosine aminotransferase,
EsAroAT1, was characterized from E. sinica. Although
they were logical candidates, neither EsAroAT1 nor EsA-
roAT2 had catalytic activity with the ephedrine alkaloid
precursors, phenylpropanedione or S-phenylacetylcarbinol.
Future studies investigating ephedrine alkaloid biosynthesis
will benefit greatly from access to an E. sinica EST data-
base developed in this study. This resource is instrumental
to the discovery of genes involved in the biosynthetic path-
way. In the long term, the availability of cloned enzymes
of ephedrine alkaloid biosynthesis may allow the develop-
ment of synthetic biology platforms for the synthesis of
amphetamine analogs with known or novel pharmaceutical
properties.
Collier RH, Kohlhaw G (1972) Nonidentity of aspartate and aromatic
aminotransferase components of transaminase-a in Escherichia-
coli. J Bacteriol 112:365–371
Cooper AJ (2004) The role of glutamine transaminase K (GTK) in
sulfur and α-keto acid metabolism in the brain, and in the possi-
ble bioactivation of neurotoxicants. Neurochem Int 44:557–577.
doi:10.1016/j.neuint.2003.12.002
Corea ORA, Ki C, Cardenas CL, Kim S-J, Brewer SE, Patten AM,
Davin LB, Lewis NG (2012) Arogenate dehydratase isoenzymes
profoundly and differentially modulate carbon flux into lignins.
J Biol Chem 287:11446–11459. doi:10.1074/jbc.M111.322164
Dal Cin V, Tieman DM, Tohge T, McQuinn R, de Vos RC, Osorio S,
Schmelz EA, Taylor MG, Smits-Kroon MT, Schuurink RC, Har-
ing MA, Giovannoni J, Fernie AR, Klee HJ (2011) Identifica-
tion of genes in the phenylalanine metabolic pathway by ectopic
expression of a MYB transcription factor in tomato fruit. Plant
Cell 23:2738–2753. doi:10.1105/tpc.111.086975
de la Torre F, De Santis L, Suarez MF, Crespillo R, Cánovas FM (2006)
Identification and functional analysis of a prokaryotic-type aspar-
tate aminotransferase: implications for plant amino acid metabo-
lism. Plant J 46:414–425. doi:10.1111/j.1365-313X.2006.02713.x
de la Torre FN, El-Azaz J, Ávila C, Cánovas FM (2014) Decipher-
ing the role of aspartate and prephenate aminotransferase activi-
ties in plastid nitrogen metabolism. Plant Physiol 164:92–104.
doi:10.1104/pp.113.232462
De-Eknamkul W, Ellis BE (1987a) Purification and characterization
of tyrosine aminotransferase activities from Anchusa officinalis
cell cultures. Arch Biochem Biophys 257:430–438
De-Eknamkul W, Ellis BE (1987b) Tyrosine aminotransferase: the
entrypoint enzyme of the tyrosine-derived pathway in rosmarinic
acid biosynthesis. Phytochemistry 26:1941–1946
Acknowledgments This research was supported by Research Grant
No. CA-9117-09 from the Canada-Israel Binational Agricultural
Research and Development Fund, the Natural Sciences and Engineer-
ing Research Council and Agriculture and Agri-Food Canada to E.L.,
P.F. and F.M. Granting agencies were not involved in study design;
in the collection, analysis and interpretation of data; in the writing of
the report; and in the decision to submit the article for publication.
K.K. was co-supervised by Norman Hüner during his M. Sc. program
(Department of Biology, University of Western Ontario). We thank
staff at the London Development and Research Centre, Agriculture
and Agri-Food Canada, Tim McDowell for assistance with GC–MS,
Sudhakar Pandurangan with RT-quantitative PCR, and Alex Mol-
nar for the preparation of figures. We acknowledge support from the
Genome Canada PhytoMetaSyn project for bioinformatic analysis.
DellaPenna D, Pogson BJ (2006) Vitamin synthesis in plants: toco-
pherols and carotenoids. Annu Rev Plant Biol 57:711–738.
doi:10.1146/annurev.arplant.56.032604.144301
Compliance with ethical standards
Ellens KW, Richardson LG, Frelin O, Collins J, Ribeiro CL, Hsieh
YF, Mullen RT, Hanson AD (2014) Evidence that glutamine
Conflict of interest The authors declare that they have no conflict
of interest.
1 3

Characterization of aromatic aminotransferases from Ephedra sinica Stapf
1219
Krizevski R, Bar E, Shalit O, Sitrit Y, Ben-Shabat S, Lewinsohn E
(2010) Composition and stereochemistry of ephedrine alkaloids
accumulation in Ephedra sinica Stapf. Phytochemistry 71:895–
903. doi:10.1016/j.phytochem.2010.03.019
Krizevski R, Bar E, Shalit OR, Levy A, Hagel JM, Kilpatrick K, Mar-
solais F, Facchini PJ, Ben-Shabat S, Sitrit Y, Lewinsohn E (2012)
Benzaldehyde is a precursor of phenylpropylamino alkaloids as
revealed by targeted metabolic profiling and comparative bio-
chemical analyses in Ephedra spp. Phytochemistry 81:71–79.
doi:10.1016/j.phytochem.2012.05.018
transaminase and omega-amidase potentially act in tandem to
close the methionine salvage cycle in bacteria and plants. Phyto-
chemistry 113:160–169. doi:10.1016/j.phytochem.2014.04.012
Finn RD, Bateman A, Clements J, Coggill P, Eberhardt RY, Eddy
SR, Heger A, Hetherington K, Holm L, Mistry J, Sonnhammer
EL, Tate J, Punta M (2014) Pfam: the protein families database.
Nucleic Acids Res 42:D222–D230. doi:10.1093/nar/gkt1223
Givan CV (1980) Aminotransferases in higher plants. In: Miflin BJ
(ed) Amino acids and derivatives, vol 5. The biochemistry of
plantsAcademic Press, New York, pp 329–357
Lee E-J, Facchini PJ (2011) Tyrosine aminotransferase contributes to
benzylisoquinoline alkaloid biosynthesis in opium poppy. Plant
Physiol 157:1067–1078. doi:10.1104/pp.111.185512
Leete E (1958) Biogenesis of d-norpseudo ephedrine in Catha-edulis.
Chem Ind 1088–1089
Gonda I, Bar E, Portnoy V, Lev S, Burger J, Schaffer AA, Tadmor
Y, Gepstein S, Giovannoni JJ, Katzir N, Lewinsohn E (2010)
Branched-chain and aromatic amino acid catabolism into aroma
volatiles in Cucumis melo L. fruit. J Exp Bot 61:1111–1123.
doi:10.1093/jxb/erp390
Graindorge M, Giustini C, Jacomin AC, Kraut A, Curien G, Matringe
M (2010) Identification of a plant gene encoding glutamate/
aspartate-prephenate aminotransferase: the last homeless enzyme
of aromatic amino acids biosynthesis. FEBS Lett 584:4357–
4360. doi:10.1016/j.febslet.2010.09.037
Grossmann K, Hutzler J, Tresch S, Christiansen N, Looser R,
Ehrhardt T (2012) On the mode of action of the herbicides cin-
methylin and 5-benzyloxymethyl-1, 2-isoxazolines: putative
inhibitors of plant tyrosine aminotransferase. Pest Manag Sci
68:482–492. doi:10.1002/ps.2319
Groves RA, Hagel JM, Zhang Y, Kilpatrick K, Levy A, Marsolais F,
Lewinsohn E, Sensen CW, Facchini PJ (2015) Transcriptome pro-
filing of khat (Catha edulis) and Ephedra sinica reveals gene can-
didates potentially involved in amphetamine-type alkaloid biosyn-
thesis. PLoS One 10:e0119701. doi:10.1371/journal.pone.0119701
Grue-Sørensen G, Spenser ID (1994) Biosynthetic route to the Ephe-
dra alkaloids. J Am Chem Soc 116:6195–6200
Lewis WH, Elvin-Lewis MPF (1977) Medical botany, plants affecting
man’s health. Wiley Interscience, USA
Lichtenthaler H (2007) Biosynthesis, accumulation and emission of
carotenoids, α-tocopherol, plastoquinone, and isoprene in leaves
under high photosynthetic irradiance. Photosynth Res 92:163–
179. doi:10.1007/s11120-007-9204-y
Liepman AH, Olsen LI (2004) Genomic analysis of aminotrans-
ferases in Arabidopsis thaliana. Crit Rev Plant Sci 23:73–89.
doi:10.1080/07352680490273419
Maeda H, Dudareva N (2012) The shikimate pathway and aromatic
amino acid biosynthesis in plants. Annu Rev Plant Biol 63:73–
105. doi:10.1146/annurev-arplant-042811-105439
Maeda H, Shasany AK, Schnepp J, Orlova I, Taguchi G, Cooper BR,
Rhodes D, Pichersky E, Dudareva N (2010) RNAi suppression
of arogenate dehydratase1 reveals that phenylalanine is synthe-
sized predominantly via the arogenate pathway in petunia petals.
Plant Cell 22:832–849. doi:10.1105/tpc.109.073247
Hagel JM, Krizevski R, Kilpatrick K, Sitrit Y, Marsolais F, Lewinsohn
E, Facchini PJ (2011) Expressed sequence tag analysis of khat
(Catha edulis) provides a putative molecular biochemical basis
for the biosynthesis of phenylpropylamino alkaloids. Genet Mol
Biol 34:640–646. doi:10.1590/S1415-47572011000400017
Hagel JM, Krizevski R, Marsolais F, Lewinsohn E, Facchini PJ (2012)
Biosynthesis of amphetamine analogs in plants. Trends Plant Sci
17:404–412. doi:10.1016/j.tplants.2012.03.004
Haller CA, Benowitz NL (2000) Adverse cardiovascular and cen-
tral nervous system events associated with dietary supplements
containing Ephedra alkaloids. N Engl J Med 343:1833–1838.
doi:10.1056/nejm200012213432502
Heilbronn J, Wilson J, Berger BJ (1999) Tyrosine aminotransferase
catalyzes the final step of methionine recycling in Klebsiella
pneumoniae. J Bacteriol 181:1739–1747
Hirata H, Ohnishi T, Ishida H, Tomida K, Sakai M, Hara M, Wata-
nabe N (2012) Functional characterization of aromatic amino
acid aminotransferase involved in 2-phenylethanol biosynthesis
in isolated rose petal protoplasts. J Plant Physiol 169:444–451.
doi:10.1016/j.jplph.2011.12.005
Maeda H, Yoo H, Dudareva N (2011) Prephenate aminotransferase
directs plant phenylalanine biosynthesis via arogenate. Nat
Chem Biol 7:19–21. doi:10.1038/nchembio.485
Mehere P, Han Q, Lemkul JA, Vavricka CJ, Robinson H, Bevan DR,
Li J (2010) Tyrosine aminotransferase: biochemical and struc-
tural properties and molecular dynamics simulations. Protein
Cell 1:1023–1032. doi:10.1007/s13238-010-0128-5
Mehta PK, Hale TI, Christen P (1993) Aminotransferases - Demonstra-
tion of homology and division into evolutionary subgroups. Eur J
Biochem 214:549–561. doi:10.1111/j.1432-1033.1993.tb17953.x
Nierop Groot MN, de Bont JA (1999) Involvement of manganese in
conversion of phenylalanine to benzaldehyde by lactic acid bac-
teria. Appl Environ Microbiol 65:5590–5593
Okada T, Mikage M, Sekita S (2008) Molecular characterization of
the phenylalanine ammonia-lyase from Ephedra sinica. Biol
Pharm Bull 31:2194–2199
Pinto JT, Krasnikov BF, Alcutt S, Jones ME, Dorai T, Villar MT, Artigues
A, Li J, Cooper AJL (2014) Kynurenine aminotransferase III and
glutamine transaminase
l are identical enzymes that have cysteine
S-conjugate β-lyase activity and can transaminate l-selenomethio-
nine. J Biol Chem 289:30950–30961. doi:10.1074/jbc.M114.591461
Prabhu PR, Hudson AO (2010) Identification and partial characteriza-
tion of an l-tyrosine aminotransferase (TAT) from Arabidopsis
thaliana. Biochem Res Int 2010:549572
Riewe D, Koohi M, Lisec J, Pfeiffer M, Lippmann R, Schmeichel
J, Willmitzer L, Altmann T (2012) A tyrosine aminotransferase
involved in tocopherol synthesis in Arabidopsis. Plant J 71:850–
859. doi:10.1111/j.1365-313X.2012.05035.x
Rothman RB, Vu N, Partilla JS, Roth BL, Hufeisen SJ, Compton-Toth
BA, Birkes J, Young R, Glennon RA (2003) In vitro charac-
terization of ephedrine-related stereoisomers at biogenic amine
transporters and the receptorome reveals selective actions as
norepinephrine transporter substrates. J Pharmacol Exp Ther
307:138–145. doi:10.1124/jpet.103.053975
Hirotsu K, Goto M, Okamoto A, Miyahara I (2005) Dual substrate rec-
ognition of aminotransferases. Chem Rec 5:160–172. doi:10.1002/
tcr.20042
Holländer-Czytko H, Grabowski J, Sandorf I, Weckermann K, Weiler
EW (2005) Tocopherol content and activities of tyrosine ami-
notransferase and cystine lyase in Arabidopsis under stress condi-
tions. J Plant Physiol 162:767–770. doi:10.1016/j.jplph.2005.04.019
Howell JM, Winstone TL, Coorssen JR, Turner RJ (2006) An evalua-
tion of in vitro protein-protein interaction techniques: assessing
contaminating background proteins. Proteomics 6:2050–2069.
doi:10.1002/pmic.200500517
Ishikawa K, Kaneko E, Ichiyama A (1996) Pyridoxal 5′-phosphate
binding of a recombinant rat serine:pyruvate/alanine:glyoxylate
aminotransferase. J Biochem 119:970–978
1 3

1220
K. Kilpatrick et al.
Rudat J, Brucher BR, Syldatk C (2012) Transaminases for the syn-
thesis of enantiopure beta-amino acids. AMB Express 2:11.
doi:10.1186/2191-0855-2-11
Yamasaki K, Sankawa U, Shibata S (1969) Biosynthesis of ephed-
rine in Ephedra. Participation of C6–C1 unit. Tetrahedron Lett
10:4099–4102
Schneider G, Kack H, LindqvistY (2000) The manifold of vitamin B6
dependent enzymes. Structure 8:R1–R6
Tohge T, Watanabe M, Hoefgen R, Fernie AR (2013) Shikimate and
phenylalanine biosynthesis in the green lineage. Front Plant Sci
4:62. doi:10.3389/fpls.2013.00062
Yoo H, Widhalm JR, Qian Y, Maeda H, Cooper BR, Jannasch AS,
Gonda I, Lewinsohn E, Rhodes D, Dudareva N (2013) An alter-
native pathway contributes to phenylalanine biosynthesis in
plants via a cytosolic tyrosine:phenylpyruvate aminotransferase.
Nat Commun 4:2833. doi:10.1038/ncomms3833
Tzin V, Malitsky S, Aharoni A, Galili G (2009) Expression of a bac-
terial bi-functional chorismate mutase/prephenate dehydratase
modulates primary and secondary metabolism associated with
aromatic amino acids in Arabidopsis. Plant J 60:156–167.
doi:10.1111/j.1365-313X.2009.03945.x
Wang C-S, Vodkin L (1994) Extraction of RNA from tissues contain-
ing high levels of procyanidins that bind RNA. Plant Mol Biol
Rep 12:132–145. doi:10.1007/bf02668374
Zerbino DR, Birney E (2008) Velvet: algorithms for de novo short
read assembly using de Bruijn graphs. Genome Res 18:821–829.
doi:10.1101/gr.074492.107
Zhang Q, Marsolais F (2014) Identification and characterization of
omega-amidase as an enzyme metabolically linked to aspara-
gine transamination in Arabidopsis. Phytochemistry 99:36–43.
doi:10.1016/j.phytochem.2013.12.020
1 3